Natural Language processing & Large Language Models | NLP & LLM

This repository is dedicated to projects and some theoretical material that I used to get into the areas of NLP and LLM in a practical/efficient way.

Index:

• How did I start?
• Courses
• AI_Agents
• Projects

Master_Thesis

My NLP journey started in 2021-2022 during the development of my master's thesis entitled:

"Learning to write medical reports from EEG data"

This aims the development of methods for clinical report writing based on EEG signals, adapting current NLP techniques and state-of-the-art captioning approaches to image, signal and video.

Due Legal/privacy restrictions, code/implementation cannot be publicly available, however the dissertation has been published and can be accessed at link.

The document includes an introdutions to NLP, state-of-the-art approach for captioning in image/signal/video, details development and comparison of 6 pipelines for EEG captioning.

Courses

Since then, interest and popularity in the area has been growing, particularly due to the emergence of transformers and LLM applications. So, in an attempt to keep up with this development, I took some courses on more recent topics in the areas offered by some of the big players. This repository includes some theoretical/notes and pratical material of thoses.

Generative AI with Large Language Models (DeepLearning.AI & AWS)

• Deeply understand generative AI, describing the key steps in a typical LLM-based generative AI lifecycle, from data gathering and model selection, to performance evaluation and deployment.
• Describe in detail the transformer architecture that powers LLMs, how they’re trained, and how fine-tuning enables LLMs to be adapted to a variety of specific use cases.
• Use empirical scaling laws to optimize the model's objective function across dataset size, compute budget, and inference requirements.
• Apply state-of-the art training, tuning, inference, tools, and deployment methods to maximize the performance of models within the specific constraints of your project.
• Challenges and opportunities that generative AI creates for businesses after hearing stories from industry researchers and practitioners.

Course Certificate: Link; More Info

Finetuning Large Language Models (DeepLearning.AI & LAMINI)

• Essential finetuning concepts and how to train a large language model.
• Understand how finetuning differs from prompt engineering, and when to use both.
• Practical experience with real data sets, and how to use techniques for projects.

Course Certificate: Link; More Info

Large Language Model Operations - LLMOps (DeepLearning.AI & Google Cloud)

• Retrieve and transform training data for supervised fine-tuning of an LLM.
• Version data and tuned models to track your tuning experiments.
• Configure an open-source supervised tuning pipeline and then execute that pipeline to train and then deploy a tuned LLM.
• Output and study safety scores to responsibly monitor and filter your LLM application’s behavior.
• Tun and deploy LLM. Practice with Tools as BigQuery data warehouse, the open-source Kubeflow Pipelines, and Google Cloud.

Course Certificate: Link; More Info

RAG - LangChain Chat with Your Data (DeepLearning.AI & LangChain)

• Retrieval Augmented Generation (RAG), a common LLM application that retrieves contextual documents from an external dataset.
• Building a chatbot that responds to queries based on the content of your documents, rather than the information it has learned in training.

Course Certificate: Link; More Info

AI_Agents

This repository explores various topics, tools, and techniques that are currently widely used in LLM (Large Language Model) projects or are gaining significant attention in the AI community. One key area of focus is AI Agents, which involve the integration of multiple LLMs to perform specific tasks collaboratively and enhance overall performance.

To do this, I used a wide variety of bibliographic sources. The list of trainings I took related to thoses topics and the related projects can be found on: Link.

Those trainings are offered by large companies and Big players in AI. Below is the list of completion certifications:

• AI Agents in LangGraph (DeepLearning.AI, LangChain & Tavily)
• Multi_AI.Agent_Systems (DeepLearning.AI, & CrewAI)
• Multi_AI.Agents_Advanced_UC (DeepLearning.AI, & CrewAI)

Disclaimer

Copyright of all materials in thoses courses belongs to DeepLearning.AI, LAMINI, AWS, Google Cloud, Tavily and LangChain and can only be used or distributed for educational purpose. You may not use or distribute them for commercial purposes.

Projects:

Here are links to my NLP & LLM projects:

• LLM-vs.-Human-Minds

  "Will you be able to win?" This game challenges humans to compete against LLM models in a variety of games.